Program Schedule

Gene Expression Profiles Discriminate Between Young Children with Human Rhinovirus (HRV) Symptomatic Infection vs Asymptomatic Detection

Session: Poster Abstract Session: Biomarkers of Immune Responses
Saturday, October 11, 2014
Room: The Pennsylvania Convention Center: IDExpo Hall BC
  • Heinonen IDweek 141006 final.pdf (2.0 MB)
  • Background:   HRVs are among the most common causes of respiratory infections in humans. However, rates of up to 50% HRV detection have been found in asymptomatic children. We evaluated the value of gene expression profiles to differentiate asymptomatic detection from symptomatic HRV infection in young children.

    Methods:  Whole blood samples from children <2 years old (n=114) were obtained at four study sites (Columbus, OH; Dallas, TX; Turku, Finland; and Malaga, Spain). Children were classified into: 1) HRV- healthy controls (HC; n=22); HRV+ asymptomatic (n=16); HRV+ outpatients (n=18); and HRV+ inpatients (n=58). RNA samples were hybridized into Illumina arrays. We used R software package and ComBat script to correct for batch effects and GeneSpring software and modular analyses for analyses.

    Results: Statistical group comparisons (p<0.05, Benjamini-Hochberg and 1.25x fold change filter) identified 402 differentially expressed genes between HC and HRV+ inpatients (HRV biosignature), which was validated in an independent group of subjects. The HRV biosignature was then applied to all study groups (114 patient samples) using unsupervised hierarchical clustering which identified 2 main patient clusters (C): C1 composed of HC and the majority of asymptomatic HRV+ subjects 15/16 (94%) and C2 with HRV+ inpatients and the majority of outpatients 15/18 (83%). Class prediction using support vector machine (SVM) identified 57 genes that classified HC vs. HRV+ inpatients with 97% accuracy and HRV+ outpatients vs. HC with 93% accuracy. Modular analysis showed a gradual increment on the up-regulation of innate immunity genes and a gradual reduction in the down-regulation of adaptive immunity genes between HRV+ inpatients, outpatients and asymptomatic children, respectively. Indeed, the latter group showed no modular differences compared with HC.

    Conclusion: Symptomatic HRV infection induced a robust and reproducible biosignature. On the other hand, transcriptional profiles of children with asymptomatic HRV+ detection were comparable to those of HC. Transcriptomic profiling represents a useful tool to discriminate between clinically irrelevant pathogen detection and true infection.

    Santtu Heinonen, MD, PhD1, Nicolas M. Suarez, PhD2, Tuomas Jartti, MD, PhD3, Silvia Oliva, MD4, Carla Garcia, MD5, Octavio Ramilo, MD6 and Asuncion Mejias, MD, PhD2, (1)Center for Vaccines and Immunity, The Research Institute at Nationwide Children's, Columbus, OH, (2)Center for Vaccines and Immunity, The Research Institute at Nationwide Children's Hospital, Columbus, OH, (3)Department of Pediatrics, Turku University Hospital, Turku, Finland, (4)Hospital Materno Infantil, Malaga's University, Malaga, Spain, (5)Division of Pediatric Infectious Diseases, UT Southwestern Medical Center, Dallas, TX, (6)Center for Vaccines and Immunity, The Research Institute At Nationwide Children's Hospital, Columbus, OH


    S. Heinonen, None

    N. M. Suarez, None

    T. Jartti, None

    S. Oliva, None

    C. Garcia, None

    O. Ramilo, None

    A. Mejias, None

    Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.

    Sponsoring Societies:

    © 2014, All Rights Reserved.

    Follow IDWeek